Uploaded image for project: 'Commons RNG'
  1. Commons RNG
  2. RNG-159

ZigguratSampler.NormalizedGaussian has incorrect Gaussian output

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Closed
    • Blocker
    • Resolution: Fixed
    • 1.4
    • 1.4
    • sampling
    • None

    Description

      The new ZigguratSampler.NormalizedGaussian does not produce suitable output from the examples sampling application:

      Compare with the current ZigguratNormalizedGaussianSampler:

      This is is repeatable with several different RNG implementations.

      This issue was not detected with the current test suite in the sampling module which uses the deciles of the distribution. The deciles may be too coarse grained to detect a problem around the distribution mean.

      Attachments

        1. pdf.gauss.modified.ziggurat.txt.png
          7 kB
          Alex Herbert
        2. pdf.gauss.ziggurat.txt.png
          7 kB
          Alex Herbert
        3. pdf.gauss.modified.ziggurat.fixed.txt.png
          7 kB
          Alex Herbert

        Issue Links

          is related to

          Improvement - An improvement or enhancement to an existing feature or task. RNG-151 Modified Ziggurat algorithm for normal and exponential sampling

          • Major - Major loss of function.
          • Closed

          Activity

            People

              Unassigned Unassigned
              aherbert Alex Herbert
              Votes:
              0 Vote for this issue
              Watchers:
              1 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: